connman.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
From: Daniel Wagner <wagi@monom.org>
To: Michael Nazzareno Trimarchi <michael@amarulasolutions.com>
Cc: connman@lists.linux.dev, Jan.Ryll@bshg.com, Simon.Holesch@bshg.com
Subject: Re: [PATCH V2] Add GetKnownServices api to connaman
Date: Thu, 28 Oct 2021 08:31:30 +0200	[thread overview]
Message-ID: <20211028063130.wwz44zuqjuoatksx@beryllium.lan> (raw)
In-Reply-To: <CAOf5uwmcmkUNO0P3mZOC94bWRpodqLeJ4kmBTSu2N0j9xw_WQQ@mail.gmail.com>

Hi Michael,

On Wed, Oct 27, 2021 at 05:25:32PM +0000, Michael Nazzareno Trimarchi wrote:
> Hi Daniel
> 
> I even have the patch to remove known-networks. I have few question:
> 
> - what should be the behavior for an ethernet on?

I might not understand you question correctly, I assumed we don't
differentiate between types. Return the complete list and let the user
decide what to do with it.

> - for wifi, it's removed and if you are connected it's disconnected is
> that correct?

I haven't tested this but I would assume nothing will happen. These
files are all considered under control of ConnMan. So if you remove
those while ConnMan is running, it wont notice it. Hmm, to get this
working correctly we might need to add a RemoveKnownNetwork call. I want
to avoid to have ConnMan monitoring the storage directory (it's fragile
stuff and difficult to get right).

If we go down the road with RemoveKnownNetwork, then I suggest the
GetKnownNetwork call returns the object path, not the storage path.

Sorry for this late enlightenment from my side. I didn't realize all the
impacts when we started the discussion.

> Apart of that:
> 
> I'm working on a patch that allow to store network configuration using
> the config storage
> but I have few questions:
> 
> - I have added autoconnect as a configuration option. I would like to
> store networks but not have them
> in auto connect mode. This should allow me to connect when they are
> visible. (ethernet??)

I can't remember all the details around AutoConnect. It's a difficult
feature as it set on the first successful connect attempt. If
configuration files are involved there are other rules.

> - I have add an option to store if I want it immutable or not. The
> stored configuration should be applied

Immutable is derived from the fact if there is a config file, the
service immutable. It doesn't really make sense to make this
configurable.

HTH,
Daniel

  reply	other threads:[~2021-10-28  6:31 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-27 14:01 [PATCH V2] Add GetKnownServices api to connaman Michael Trimarchi
2021-10-27 17:25 ` Michael Nazzareno Trimarchi
2021-10-28  6:31   ` Daniel Wagner [this message]
2021-10-28  6:16 ` Daniel Wagner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20211028063130.wwz44zuqjuoatksx@beryllium.lan \
    --to=wagi@monom.org \
    --cc=Jan.Ryll@bshg.com \
    --cc=Simon.Holesch@bshg.com \
    --cc=connman@lists.linux.dev \
    --cc=michael@amarulasolutions.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).